mobilevit-xx-small-finetuned-eurosat
This model is a fine-tuned version of apple/mobilevit-xx-small on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1926
- Accuracy: 0.9507
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5074 | 1.0 | 190 | 1.3433 | 0.7078 |
| 0.9398 | 2.0 | 380 | 0.7177 | 0.85 |
| 0.7035 | 3.0 | 570 | 0.4252 | 0.9070 |
| 0.5435 | 4.0 | 760 | 0.3080 | 0.9281 |
| 0.5007 | 5.0 | 950 | 0.2465 | 0.9389 |
| 0.4533 | 6.0 | 1140 | 0.2291 | 0.9444 |
| 0.3961 | 7.0 | 1330 | 0.1991 | 0.9496 |
| 0.3949 | 8.0 | 1520 | 0.1926 | 0.9507 |
| 0.4302 | 9.0 | 1710 | 0.1928 | 0.95 |
| 0.4061 | 10.0 | 1900 | 0.1931 | 0.9463 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.1
- Tokenizers 0.13.3